ModaLink: Unifying Modalities for Efficient Image-to-PointCloud Place Recognition.
Weidong XieLun LuoNanfei YeYi RenShaoyi DuMinhang WangJintao XuRui AiWeihao GuXieyuanli ChenPublished in: CoRR (2024)
Keyphrases
- image analysis
- image data
- multiscale
- image features
- image representation
- image collections
- single image
- image classification
- template matching
- image content
- hough transform
- input image
- low level
- grey level
- image retrieval
- pixel values
- lighting conditions
- similarity measure
- image matching
- image segmentation
- aerial images
- image pixels
- image processing
- image noise
- region of interest
- vector field
- image regions
- keypoints
- segmentation method
- segmentation algorithm
- post processing
- high resolution
- image set
- visual data
- object recognition